Digital Twin for Advanced Network Planning: Tackling Interference

TL;DR

This paper presents a Digital Twin framework utilizing machine learning and clustering to enhance network planning and interference detection, reducing manual effort and improving operational efficiency in next-generation wireless networks.

Contribution

It introduces a novel Digital Twin-based methodology for RF planning and interference analysis, integrating clustering techniques for anomaly detection in operational networks.

Findings

Improved network planning accuracy through data-driven anomaly detection

Effective interference analysis using clustering techniques

Simulation results show enhanced operational efficiency

Abstract

Operational data in next-generation networks offers a valuable resource for Mobile Network Operators to autonomously manage their systems and predict potential network issues. Machine Learning and Digital Twin can be applied to gain important insights for intelligent decision-making. This paper proposes a framework for Radio Frequency planning and failure detection using Digital Twin reducing the level of manual intervention. In this study, we propose a methodology for analyzing Radio Frequency issues as external interference employing clustering techniques in operational networks, and later incorporating this in the planning process. Simulation results demonstrate that the architecture proposed can improve planning operations through a data-aided anomaly detection strategy.